@xxx.edu Fri Jan 6 12:26:01 1995 Date: Fri, 6 Jan 95 09:18:07 GMT @xxx.edu @xxx.za Subject: Majordomo file: list 'diy_efi' file 'archive_num_32' -- >From owner-diy_efi Thu Oct 6 19:04:05 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA13023; Thu, 6 Oct 94 19:04:05 GMT Received: from hwking.cca.rockwell.com by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA13018; Thu, 6 Oct 94 15:03:54 -0400 Received: by hwking.cca.rockwell.com (5.65/DEC-Ultrix/4.3) id AA21749; Thu, 6 Oct 1994 14:03:51 -0500 Received: by ohura.cca.rockwell.com (5.65/DEC-Ultrix/4.3) id AA03624; Thu, 6 Oct 1994 14:03:49 -0500 @xxx.com> To: DIY_EFI Subject: Re: FW: Processor selection for datalogger In-Reply-To: Your message of "Thu, 06 Oct 94 08:11:00 EDT." @xxx.org> Date: Thu, 06 Oct 94 14:03:48 -0500 @xxx.com X-Mts: smtp Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI >I would like to build a data logger to monitor a few engine >parameters. No fair. That's my idea! ;) >Does anyone have a feel for what the injector PW signal looks like >(timing)? I suspect it is a pulse between 1-10msec wide at idle. I have some scms. from a friend's home-built datalogger, and he measures from 1ms to 40ms. The application is a '87 Buick GN.(C3 ECM) It's an interesting circuit, as he uses _no_ microprocessors. >The technical service manual for my car describes the sequence and the >crank angle at which the injection takes place, but it doesn't tell >you how wide the pulses are. Check out the section on injector balancing. It might give you a good feel for the pulse width. >How often do I need to sample to get a reasonable reading (i.e. is it >5msec +/- 0.000001msec, needing a 1e6 counter, or is it 1-100msec, >etc.)? I realize this varies from car to car, but I am only looking >for an ``order of magnitude'' starting point. The GN datalogger uses 2 74HC590s. I'm looking at using a 68HC711G5, (if I can find one) and using the input capture channels (6 possible on the G5) to measure this and all my other PW signals. My concern is turning to sampling rate now, too, and I haven't really dug into the systems issues. Let me know what you find out. Dig @xxx.com Syclone/Typhoon mailing list. Feel the power of the wind. >From owner-diy_efi Fri Oct 7 04:20:14 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA14546; Fri, 7 Oct 94 04:20:14 GMT Received: from knuth.mtsu.edu by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA14541; Fri, 7 Oct 94 00:20:09 -0400 Received: by knuth.mtsu.edu (Smail3.1.28.1 #21) id m0qt5hh-000CuTC; Thu, 6 Oct 94 22:10 CDT @xxx.edu> @xxx. Lusky) Subject: Re: FW: Processor selection for datalogger To: DIY_EFI Date: Thu, 6 Oct 1994 22:10:41 -0500 (CDT) @xxx. Landrau" at Oct 6, 94 08:11:00 am X-Mailer: ELM [version 2.4 PL24alpha3] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1423 Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI Roberto L. Landrau writes: > > Does anyone have a feel for what the injector PW signal looks like > (timing)? I suspect it is a pulse between 1-10msec wide at idle. > The technical service manual for my car describes the sequence and the > crank angle at which the injection takes place, but it doesn't tell > you how wide the pulses are. With latemodel GM (Rochester, i think) injectors (the ball type with the director plate) you can go down a little bit below 1ms if youve got a really good controller. I think most people like to stay above 2ms. > My specific question is: how wide is the injector pulse signal? How > often do I need to sample to get a reasonable reading (i.e. is it > 5msec +/- 0.000001msec, needing a 1e6 counter, or is it 1-100msec, > etc.)? I realize this varies from car to car, but I am only looking > for an ``order of magnitude'' starting point. An injector fires once every two engine revolutions. It needs at least 1ms offtime between firing. Maximum dutycycle is about 90% I believe. So your frequency is a function of RPM, then you apply both of the above rules to figure out what your maximum pulse width could be. -- @xxx.edu http://www.edge.net/erc/lusky.html (615) 455-9915 ------------------------------------ ------------------------------ 68 Camaro Convertible - 350 / TH350 \_/ 80 Toyota Celica - 20R / 5spd >From owner-diy_efi Fri Oct 7 05:10:29 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA14629; Fri, 7 Oct 94 05:10:29 GMT Received: from maxwell.ee.washington.edu by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA14623; Fri, 7 Oct 94 01:10:27 -0400 Received: by maxwell.ee.washington.edu (1.37.109.4/UW-NDC Revision: 2.26 ) id AA23621; Thu, 6 Oct 94 22:10:18 -0700 @xxx.edu> @xxx.edu> Subject: Re: FW: Processor selection for datalogger To: DIY_EFI Date: Thu, 6 Oct 94 22:10:18 PDT @xxx. Lusky" at Oct 6, 94 10:10 pm Mailer: Elm [revision: 70.85] Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI Hi: A good way to estimate the maximum pulse width your injectors will see relies on the following two facts... 1. The most fuel injected into the cylinder will be at maximum torque rpm. You know what that rpm will be. 2. As mentioned in Jonathan Luskeys post, the duty cycle of the injector has an upper bound of less than 100%. He said 90%, I've heard 80%. It's in that ballpark, anyway. So the widest pulse occurring would be roughly approximated by .85*(2/max_torque_rpm). This is all IMHO, and intended as a way to *estimate* the pulse width. I hope it helps. Mike Gruber >From owner-diy_efi Fri Oct 7 05:52:03 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA14688; Fri, 7 Oct 94 05:52:03 GMT Received: from maxwell.ee.washington.edu by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA14683; Fri, 7 Oct 94 01:52:01 -0400 Received: by maxwell.ee.washington.edu (1.37.109.4/UW-NDC Revision: 2.26 ) id AA29727; Thu, 6 Oct 94 22:51:52 -0700 @xxx.edu> @xxx.edu> Subject: Re: FW: Processor selection for datalogger To: DIY_EFI Date: Thu, 6 Oct 94 22:51:51 PDT @xxx.edu>; from "Mike Gruber" at Oct 6, 94 10:10 pm Mailer: Elm [revision: 70.85] Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI Earlier, I mistakenly wrote: > approximated by .85*(2/max_torque_rpm). > The approximation *should be* .85*(2/(max_torque_rpm/60)), since we want the answer in seconds, not minutes. Sorry for the inaccuracy. > > Mike Gruber > > >From owner-diy_efi Mon Oct 10 06:03:05 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA23212; Mon, 10 Oct 94 06:03:05 GMT Received: from eigen.ee.ualberta.ca by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA23207; Mon, 10 Oct 94 02:02:56 -0400 @xxx.edu> Received: by eigen.ee.ualberta.ca (1.37.109.4/15.6) id AA19124; Mon, 10 Oct 94 00:02:48 -0600 @xxx.ca> Subject: Re: P4 GM ECM using 68HC11 To: DIY_EFI Date: Mon, 10 Oct 94 0:02:47 MDT @xxx.edu" at Oct 6, 94 11:26 am Mailer: Elm [revision: 70.85] Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI > > What cars was the P4 used in? Is it in all the TPI camaros/firebirds? > Thats the only GM ECM I've seen using a 27256. It would be great to > tear one of those down and figure it out. > > Is the OEM code on ROM on the chip? I know you can get 68hc11 mask > programmed at motorolla, but considering there is an external 32k > EPROM maybee they don't use that. Masked programming is only done if you're making millions of the same thing. Most cars used the P4 ECM (post 1986). Older cars and most trucks (up to about 1991) used the C3 which is very old... Yes, the PROM contains code. About 20k worth of it. Have fun!!! -Dale >From owner-diy_efi Mon Oct 10 13:03:32 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA23752; Mon, 10 Oct 94 13:03:32 GMT Received: from gw1.att.com by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA23746; Mon, 10 Oct 94 09:03:21 -0400 Received: from uscbu.ih.att.com by ig1.att.att.com id AA17386; Mon, 10 Oct 94 09:01:51 EDT Received: by uscbu.ih.att.com (4.1/EMS-1.1.1 SunOS) id AA13266; Mon, 10 Oct 94 08:02:28 CDT @xxx.com Received: from usgp1.ih.att.com by uscbu.ih.att.com (4.1/EMS-1.1.1 SunOS) id AA13244; Mon, 10 Oct 94 08:02:23 CDT Received: by usgp1.ih.att.com (5.0/EMS-1.1 Sol2) id AA07221; Mon, 10 Oct 1994 08:02:37 +0600 Date: Mon, 10 Oct 1994 08:02:37 +0600 @xxx.com> @xxx.com (Bohdan L Bodnar) To: DIY_EFI Subject: Re: P4 GM ECM using 68HC11 Content-Type: text Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI Partially correct, Dale. Computer Command Control is a SYSTEM, not just hardware. The GEN V computer which will eventually make it into production vehicles is still C3. Yes, the basic C3 is old (1981 for full function systems, 1982 for Min-T systems). Regards, Bohdan Bodnar @xxx.com >From owner-diy_efi Tue Oct 11 01:34:41 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA26510; Tue, 11 Oct 94 01:34:41 GMT Received: from shiva.trl.oz.au by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA26505; Mon, 10 Oct 94 21:34:32 -0400 Received: by shiva.trl.OZ.AU id AA12434 @xxx.edu); Tue, 11 Oct 1994 11:34:23 +1000 @xxx.au> @xxx.AU> Subject: Injector timing qns. To: DIY_EFI Date: Tue, 11 Oct 1994 11:34:22 +1000 (EST) X-Mailer: ELM [version 2.4 PL20] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 964 Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI Hi there, I've got a few questions regarding injector timing: i/ Does the timing of the injector relative to the timing of the intake opening matter much?. IE if you've got the same injector on time, but change the 'phase angle' relative to the cam, will the power output vary much? Rewording, will power output change much between full sequential and simultaneous injection, if you assume (as with L-Jetronic) that the fuel can "Hang around" in the intake while the valve is shut, or for optimum power should the fuel injector only be turned on while air is flowing into the engine? ii/ Does anyone have a graph of fuel flow/minute vs injector duty cycle at high RPM. (Concerning injector opening&closing times) I'm in the process of building a basic analog injection computer so I can get a feel for how the fuel algorithms work, then put it into a more elegant microcontroller form (Probably 8051 based) Cheers, Craig. >From owner-diy_efi Tue Oct 11 02:11:53 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA26652; Tue, 11 Oct 94 02:11:53 GMT Received: from knuth.mtsu.edu by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA26647; Mon, 10 Oct 94 22:11:36 -0400 Received: by knuth.mtsu.edu (Smail3.1.28.1 #21) id m0quWgp-000CvfC; Mon, 10 Oct 94 21:11 CDT @xxx.edu> @xxx. Lusky) Subject: Re: Injector timing qns. To: DIY_EFI Date: Mon, 10 Oct 1994 21:11:43 -0500 (CDT) @xxx.AU> from "Craig Pugsley" at Oct 11, 94 11:34:22 am X-Mailer: ELM [version 2.4 PL24alpha3] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1749 Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI Craig Pugsley writes: > > i/ Does the timing of the injector relative to the timing of > the intake opening matter much?. IE if you've got the same > injector on time, but change the 'phase angle' relative to the > cam, will the power output vary much? At lower injector pulse widths it can make a noticeably difference in fuel consumption and emissions. At higher pulsewidths it doesn't have any impact. > Rewording, will power output change much between full sequential > and simultaneous injection, if you assume (as with L-Jetronic) > that the fuel can "Hang around" in the intake while the valve > is shut, or for optimum power should the fuel injector only be > turned on while air is flowing into the engine? At WOT sequential injection is worthless IMHO. Your injectors will be spending considerably more time open than the intake valves. > ii/ Does anyone have a graph of fuel flow/minute vs injector duty > cycle at high RPM. (Concerning injector opening&closing times) This is going to be a function of fuel pressure and different for every model of injector. > I'm in the process of building a basic analog injection computer so I > can get a feel for how the fuel algorithms work, then put it into a more > elegant microcontroller form (Probably 8051 based) Doesn't sound particulary useful to me... I think it'd be much more useful to put together a good little data logging system and then sample an OEM system in action. -- @xxx.edu http://www.edge.net/erc/lusky.html (615) 455-9915 ------------------------------------ ------------------------------ 68 Camaro Convertible - 350 / TH350 \_/ 80 Toyota Celica - 20R / 5spd >From owner-diy_efi Tue Oct 11 03:04:19 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA27929; Tue, 11 Oct 94 03:04:19 GMT Received: from shiva.trl.oz.au by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA27920; Mon, 10 Oct 94 23:04:05 -0400 Received: by shiva.trl.OZ.AU id AA18450 @xxx.edu); Tue, 11 Oct 1994 13:03:58 +1000 @xxx.au> @xxx.AU> Subject: Re: Injector timing qns. To: DIY_EFI Date: Tue, 11 Oct 1994 13:03:57 +1000 (EST) @xxx. Lusky" at Oct 10, 94 09:11:43 pm X-Mailer: ELM [version 2.4 PL20] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1850 Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI Hi there, > > > Rewording, will power output change much between full sequential > > and simultaneous injection, if you assume (as with L-Jetronic) > > that the fuel can "Hang around" in the intake while the valve > > is shut, or for optimum power should the fuel injector only be > > turned on while air is flowing into the engine? > > At WOT sequential injection is worthless IMHO. Your injectors will > be spending considerably more time open than the intake valves. That's basically what I was wanting to know. The main aim at lower RPM is to get more torque (Mazda rotary = poor low down torque) > > ii/ Does anyone have a graph of fuel flow/minute vs injector duty > > cycle at high RPM. (Concerning injector opening&closing times) > > This is going to be a function of fuel pressure and different for > every model of injector. Hmm... I kind of mis-worded the question. What I'd like to know is how linear the fuel flow vs injector time is (For a given fuel pressure) when injector time is small, and if compensation for this deviation is nessecary. > > I'm in the process of building a basic analog injection computer so I > > can get a feel for how the fuel algorithms work, then put it into a more > > elegant microcontroller form (Probably 8051 based) > > Doesn't sound particulary useful to me... I think it'd be much more > useful to put together a good little data logging system and then sample > an OEM system in action. I've already done some measurements of an ECU (aftermarket) and I'm trying to put it into practice. I feel an analog system is easier to "bash into shape" than microprocessor code, as I have only limited micro programming experience. I have enough readings to make a fuel map, but don't have enough info about how other inputs should affect the injector time. Regards, Craig. >From owner-diy_efi Tue Oct 11 03:30:14 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA28599; Tue, 11 Oct 94 03:30:14 GMT Received: from knuth.mtsu.edu by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA28588; Mon, 10 Oct 94 23:30:03 -0400 Received: by knuth.mtsu.edu (Smail3.1.28.1 #21) id m0quXun-000Cv4C; Mon, 10 Oct 94 22:30 CDT @xxx.edu> @xxx. Lusky) Subject: Re: Injector timing qns. To: DIY_EFI Date: Mon, 10 Oct 1994 22:30:13 -0500 (CDT) @xxx.AU> from "Craig Pugsley" at Oct 11, 94 01:03:57 pm X-Mailer: ELM [version 2.4 PL24alpha3] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1228 Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI Craig Pugsley writes: > > That's basically what I was wanting to know. The main aim at lower RPM > is to get more torque (Mazda rotary = poor low down torque) Ugh, how many times are you planning to fire the injector each crank revolution? > Hmm... I kind of mis-worded the question. What I'd like to know is > how linear the fuel flow vs injector time is (For a given > fuel pressure) when injector time is small, and if compensation for > this deviation is nessecary. It depends on the injectors, your fuel pressure, and how low you go. On GM speed density systems this is compensated for by fudging the volumetric efficiency tables.... sorta beat it to fit kinda thing. Aftermarket speed density systems are similar, except that they generally look up pulse width directly from (RPM,MAP) instead of pretending to go through a VE step. Bottom line: compensation is done in the calibration, the algorithms for speed density assume the injectors are linear. -- @xxx.edu http://www.edge.net/erc/lusky.html (615) 455-9915 ------------------------------------ ------------------------------ 68 Camaro Convertible - 350 / TH350 \_/ 80 Toyota Celica - 20R / 5spd >From owner-diy_efi Tue Oct 11 15:54:47 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA13389; Tue, 11 Oct 94 15:54:47 GMT Received: from [192.150.149.1] by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA13384; Tue, 11 Oct 94 11:54:43 -0400 Received: from cpu.us.dynix.com by amlibs.com with smtp Sender: owner-diy_efi Precedence: bulk @xxx.m0qujZE-0007YXC;Tue (Linux Smail3.1.28.1 #5), 11.Oct.94.09:56.MDT Received: by cpu.us.dynix.com (AIX 3.2/UCB 5.64/4.03) id AA18937; Tue, 11 Oct 1994 09:54:02 -0600 Date: Tue, 11 Oct 1994 09:37:23 -700 (MDT) @xxx.com> Subject: Re: Injector timing qns. To: DIY_EFI @xxx.edu> @xxx.com> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Mon, 10 Oct 1994, Jonathan R. Lusky wrote: > Bottom line: compensation is done in the calibration, > the algorithms for speed density assume the injectors are linear. In BOSCH systems, there are TWO adjustments to the final pulse width 1) For voltage, to compensate for different opening/closing times 2) A final trim, which corrects extreme (large or small) pulse widths to linearize the "squirt" function .. They are contained in simple 2D tables ... "maps" if you will .. Jim Conforti @xxx.com> ÿ